GETTING IT RIGHT VACCINE EVIDENCE AND EDITORIAL DOSSIER Research checked 6 October 2026. Prepared for a broad-audience factual website. RECOMMENDATION Lead with a narrowly defined measles/MMR page. The strongest short sign is: “Two doses of MMR are about 97% effective against measles.” Pair it with: “What the evidence shows, including the risks.” Avoid “Vaccines are safe” as a standalone sign: audiences can reasonably interpret it as a guarantee of no harm, and it hides differences between products and people. The quantitative sign concerns measles, not mumps, every vaccine, or lifelong immunity in every person. Use three separate sections: (1) how vaccine evidence works; (2) measles and MMR; (3) COVID vaccines. Give autism its own evidence explainer linked from the MMR page. Do not make COVID uncertainty spill over into MMR, or use MMR effectiveness to imply equivalent COVID infection prevention. This is an editorial recommendation, not a medical recommendation for an individual. FIRST PAGE DRAFT Two doses of MMR are about 97% effective against measles. MMR protects against measles, mumps and rubella. Measles protection is strong, but not perfect: some vaccinated people can still get infected. The 97% estimate describes a reduction in measles risk compared with people who are unvaccinated under comparable conditions. It does not mean every exposed vaccinated person has a 3% chance of infection. [1] Measles can cause pneumonia, brain inflammation and death. Its risks vary with age, health and access to care. Vaccination also has risks. MMR commonly causes short-lived symptoms such as fever or soreness; serious reactions are rare. A small extra risk of fever-related seizures occurs after MMR, especially in young children. The combined MMRV product has a different first-dose seizure risk from MMR and a separate chickenpox shot. [2,3] Large studies and systematic reviews do not support a causal connection between MMR and autism. That conclusion is about a specific proposed harm, not proof that MMR has no side effects. [4,5] MMR is a live vaccine and is unsuitable for some people, including during pregnancy and for certain severe immune deficiencies. Current eligibility and timing depend on local guidance and individual circumstances. [6] Suggested page sections: The short answer; What “97% effective” means; Benefits and known risks; What studies say about autism; Who the evidence covers; Sources and what could change this page. Put limitations beside the relevant number, not in an invisible footer. CORE EVIDENCE CARDS [1] CDC, Measles Vaccination. Current page retrieved 6 October 2026. https://www.cdc.gov/measles/vaccines/ Authority guidance, full webpage inspected. Estimates: one MMR dose 93% effective against measles; two doses 97%. These are established summaries, not a new 2026 trial. The estimate is disease-specific; mumps estimates are lower. Use “about” and label the endpoint. For calculation teaching, if comparable unvaccinated risk were hypothetically 10%, 97% relative effectiveness would imply 0.3% risk, not 3%. Label this an illustration rather than observed risk. [2] CDC, MMR Vaccine Safety, 31 July 2024, retrieved 6 October 2026. https://www.cdc.gov/vaccine-safety/vaccines/mmr.html Full authority webpage inspected. Fever-related seizures: approximately one additional case per 3,000–4,000 vaccinated children under seven, around 6–14 days later. Temporary low platelets: one study estimated approximately 1 per 40,000 vaccinated children within six weeks. Severe allergy is rare. For first doses in children 12–23 months, MMRV approximately doubles short-window febrile-seizure risk versus separate MMR plus varicella. These are different comparisons: do not mix attributable MMR seizure risk with all observed events after two separate shots. Do not describe febrile seizures as epilepsy. Do not round rare risks to zero. [3] CDC Yellow Book 2026, Measles. https://www.ncbi.nlm.nih.gov/books/NBK620906/ Authority chapter indexed text inspected. Measles complications include diarrhea, ear infection, pneumonia and rare delayed neurological disease. Complication denominators are people with measles; vaccine adverse-event denominators are doses or vaccinated children. The two are not directly comparable without the chance of infection over a specified time. Avoid a single undifferentiated “disease risk versus shot risk” chart. Rates depend on setting and age. [4] Di Pietrantonj et al. Cochrane 2021, Vaccines for measles, mumps, rubella, and varicella in children. DOI 10.1002/14651858.CD004407.pub5. https://pmc.ncbi.nlm.nih.gov/articles/PMC8607336/ https://www.cochrane.org/evidence/CD004407_does-measles-mumps-rubella-and-varicella-mmrv-vaccine-protect-children-and-does-it-cause-harmful Review: 138 studies and 23,480,668 participants across effectiveness and safety outcomes, not 23 million children in an autism experiment. Population: children up to 15. Review used independent bias assessment, design-specific tools, random-effects pooling and GRADE; 53 studies had low, 55 unclear and 30 high risk of bias. Moderate certainty for MMR effectiveness. Supports protection and no autism association while detecting seizures and thrombocytopenia. Important limitation: heterogeneous designs/products/strains and publication bias cannot be excluded. The direct PMC open encountered a browser check, but indexed full-text methods and review summary were accessible and inspected; no claim that every included primary paper was read. [5] Hviid et al. 2019, MMR Vaccination and Autism: A Nationwide Cohort Study. DOI 10.7326/M18-2101. https://pubmed.ncbi.nlm.nih.gov/30831578/ Abstract and indexed summary only in this pass; full article not independently audited. Danish registry cohort: 657,461 children born 1999–2010, follow-up through August 2013; adjusted autism hazard ratio 0.93, 95% CI 0.85–1.02 for MMR vaccinated versus unvaccinated. No increased risk supported. Observational design does not mathematically exclude every minute effect or possible subgroup; do not interpret a point estimate below one as proof MMR prevents autism. Use alongside the systematic review, not alone. [6] CDC Pink Book, Measles, current webpage. https://www.cdc.gov/pinkbook/hcp/table-of-contents/chapter-13-measles.html Authority chapter; full contraindications and pregnancy sections inspected. Pregnancy and severe immunocompromise are listed contraindications. Chapter itself warns that schedule recommendations may be outdated; use it for the specific verified safety facts, and recheck live recommendations at publication. A public explanatory page should direct personal scheduling questions to local guidance and a qualified clinician rather than prescribe timing. AUTISM PAGE Suggested sign: “The evidence does not support a vaccine–autism link.” For an MMR-specific sign, “MMR does not cause autism” reflects the established conclusion; accompany it with study scope rather than an impossible zero-risk claim about all outcomes. [7] WHO, Vaccines, thiomersal and autism spectrum disorder: evidence review 2010–2025, publication landing page dated 3 September 2026. https://www.who.int/publications/i/item/9789240123533 Current publication overview inspected; final 40-page publication download failed in this pass. Do not claim final-version full-method review. It reports that higher-quality studies do not support causation; positive studies were low-strength/high-bias. It covers studies through August 2025, despite its 2026 publication date. [8] WHO commissioned evidence review released December 2025, full technical PDF. https://cdn.who.int/media/docs/default-source/medicines/pharmacovigilance/rev1_vaccines-autism-and-asd-evidence-review-2010-2025_12-dec7ab84d63-5110-4b2c-9882-19651446b8cd.pdf?download=true&sfvrsn=daab5374_6 Full PDF text available; methods inspected. Standardized independent extraction by three reviewers; at least two assessed evidence strength and bias, drawing on MetaQAT, RoB 2, Newcastle–Ottawa and causality principles. This is a qualitative synthesis, not one pooled randomized estimate. Keep December technical and September final publication versions distinct until differences are checked. Do not count them as independent replication. COVID PAGE Suggested sign: “COVID vaccines reduce severe illness. Benefits and risks vary.” This needs a dated explanatory page, especially because age, prior immunity, product and circulating virus matter. Do not lead with a universal numerical benefit or “safe for everyone.” [9] Wiegand et al. JAMA Network Open, 23 June 2026, Interim Estimated Effectiveness of 2025–2026 COVID-19 Vaccines in Adults Using a Test-Negative Design. DOI 10.1001/jamanetworkopen.2026.25152. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2850668 Full article, methods/results/limitations inspected. US September–December 2025; immunocompetent adults ≥18; 85,725 emergency/urgent-care encounters, 26,073 hospitalizations. Compared with no 2025–26 dose, adjusted effectiveness was 50% (42–57%) for COVID-associated emergency/urgent care, 55% (41–66%) for hospitalization, median 46–47 days after dose. These are incremental benefits over existing immunity, not versus immunologically naive people. Test-negative observational estimates can retain confounding and misclassification; insufficient power for several subgroups, critical illness and waning. Not evidence of 55% protection for every age, a full year, death, infection or a 2026–27 product. Do not derive absolute risk reduction from the case/control proportions. [10] FDA safety communication, 25 June 2025. https://www.fda.gov/vaccines-blood-biologics/safety-availability-biologics/fda-approves-required-updated-warning-labeling-mrna-covid-19-vaccines-regarding-myocarditis-and Full communication inspected. Commercial-claims estimates for 2023–24 mRNA formulations: unadjusted myocarditis and/or pericarditis incidence during days 1–7 was approximately 8 per million doses for ages 6 months–64 years and 27 per million in males 12–24. These are observed unadjusted incidences, not background-subtracted excess cases. Do not substitute them for original-series, second-dose, booster, different-product or different-window risks. FDA describes persistent MRI abnormalities in some myocarditis patients, with uncertain long-term significance. Avoid “always mild” or “everyone fully recovers.” The underlying claims analysis and long-term outcome dataset were not independently reanalyzed. [11] Patone et al. Circulation 2022;146:743–754. DOI 10.1161/CIRCULATIONAHA.122.059970. https://pmc.ncbi.nlm.nih.gov/articles/PMC9439633/ https://pubmed.ncbi.nlm.nih.gov/35993236/ Indexed full-text methods/limitations inspected after direct PMC browser check. England 2020–21, people ≥13 who ultimately received vaccination; self-controlled case series, calendar-adjusted 1–28-day myocarditis admission/death windows. In males under 40, excess events per million were 97 (91–99) after second-dose mRNA-1273 versus 16 (12–18) following a positive test before vaccination. This refutes a blanket claim that infection-associated myocarditis is always higher. Limitations: hospital/death coding, first events only, sparse adolescent events, undetected infections, and exclusion of people dying before vaccination. Historical formulation/variant era; not a current all-product comparison or overall benefit–risk calculation. Additional editorial safeguards: Vaccination and infection are not mutually exclusive choices. Infection has outcomes beyond myocarditis; vaccination can change subsequent infection/severity risk. Benefit–risk needs a defined future horizon and exposure probability. Prior infection can reduce baseline risk without making additional vaccination automatically useless; marginal benefit and timing still vary. Separate myocarditis, pericarditis, clotting syndromes, anaphylaxis and nonspecific symptoms rather than summing unlike definitions. Do not call protein vaccines free of myocarditis risk without checking their product labels. A complete product-by-product adverse-event inventory remains a publication-stage task. CURRENT POLICY IS NOT IDENTICAL TO EVIDENCE [12] CDC live 2026–27 guidance, updated 23 September 2026, retrieved 6 October. https://www.cdc.gov/covid/hcp/vaccine-considerations/routine-guidance.html Recommends 2026–27 vaccination for adults ≥18 and immunocompromised children ≥6 months; other children through shared clinical decision-making. It explicitly distinguishes recommendations from FDA indications and identifies some use as off-label. Search snippets still returned superseded 2025–26 policy: always open the page. Do not turn this dossier into a dosing guide. Recheck at launch. [13] WHO COVID vaccine Q&A, 15 September 2026. https://www.who.int/news-room/questions-and-answers/item/coronavirus-disease-(covid-19)-vaccines Current full authority webpage inspected. Recommendations prioritize severe-disease risk and national context; routine revaccination of healthy children/adolescents is not recommended, while high-risk populations and pregnancy receive specific attention. Policy differences reflect population risk, program feasibility and priorities as well as evidence. The site should label country, organization, date, recommendation versus authorization, and link to the current source. Do not present WHO priorities as US eligibility rules. VACCINES GENERALLY [14] Shattock et al. Lancet 2024;403:2307–2316. DOI 10.1016/S0140-6736(24)00850-X. https://pmc.ncbi.nlm.nih.gov/articles/PMC11140691/ https://pubmed.ncbi.nlm.nih.gov/38705159/ Indexed full methods and results inspected. Models covering 14 pathogens, 194 countries and 1974–2024 estimated 154 million deaths averted. Framework combined 22 models with extrapolation/imputation, comparing vaccination history with a hypothetical no-vaccination world. Useful broad context; label it a modelled estimate, not counted individual rescues. This study excludes COVID vaccination and is not a safety analysis. Avoid “154 million proves every shot benefits every person.” GENERAL EVIDENCE LITERACY Recommended editorial rules: distinguish temporal reports from causal adverse reactions; explain that surveillance signals prompt investigation. Preserve both relative and absolute measures where baseline risk is credible. State uncertainty intervals, outcome definition, comparator, age/sex strata, geography, product/dose and observation window. A denominator of doses is not interchangeable with people, infections or person-years. Distinguish controlled trials, observational associations, systematic reviews and models. Do not interpret failure to reject a null as proof of exact equivalence, or an imprecise subgroup as proof of no effect. Explain “safe” as an assessed benefit–risk balance for a specified use, with known risks and continued surveillance, rather than absence of any possible harm. Avoid stigmatizing parents, autistic people, people with adverse reactions, or people asking questions. Invite inspection of methods and corrections. Acknowledge real safety signals early; do not bury them behind reassurance. RELEASE CHECKLIST AND UPDATE TRIGGERS Before publication: have a clinically qualified reviewer check all medical copy, obtain and compare the final WHO review, inspect current product labels for the specific COVID formulations named, and verify every source link. Mark abstract-only cards explicitly. Avoid raw outbreak counts unless a dated live-data workflow exists. Reopen a claim when a credible replicated study or systematic review changes its estimate; a regulator confirms a new signal, changes a label or withdraws a product; a formulation/dose schedule changes; circulating variants or waning alter the relevant endpoint; follow-up clarifies myocarditis outcomes; local eligibility guidance changes; or an erratum/retraction affects a core source. Keep an accessible correction history and preserve the version a printed QR code originally supported. This dossier supports the MMR launch claim now. It does not constitute an exhaustive review of every vaccine, COVID product or subgroup, and supplies no personalized medical advice.